Multnomah County Adolescent Alcohol

1
Multnomah County Adolescent Alcohol Drug
Treatment System 1995-2001

• Presentation for the Multnomah County
• Department of Community Justice
• August 2002

2
The presentation is divided into two parts.

• Part 1 A view of Multnomah Countys adolescent
  alcohol and drug treatment system based on data
  from the States Client Process Monitoring System
  (CPMS).
• Part 2 Client characteristics and flow into and
  through treatment based on data collected under
  the Targeted Capacity Expansion grant to enhance
  services for youth.

3
Part 1 A view based on State CPMS dataSome
Definitions
State CPMS Data
Youth AD Treatment System
4
State CPMS Data

• Client Process Monitoring System -- State
  operated, mainframe-based relational database.
• Includes all treatment episodes (services)
  provided by agencies who receive funds from the
  State, regardless of who ultimately pays for the
  service.
• Based on forms, (originally paper but
  increasingly electronic) completed by treatment
  provider agencies and submitted to the State.
• Generally blocked into fiscal years -- July 1
  thru June 30.

5
A word about the AD CPMS data ...

• The database supports three service systems --
  mental health (MH), developmental disabilities
  (DD) and alcohol and drug treatment (AD).
• While others have found distinct problems with
  the MH and DD portions of the system, studies of
  the AD data suggest that while there may be
  coverage problems (i.e., some episodes may not be
  recorded), the integrity of the data is, for the
  most part, high.
• The integrity of the data is highest when the
  treatment episode is the unit of analysis.

6
Youth AD Treatment System

• Youth includes treatment episodes in which the
  client is under age 18 at the time of enrollment.
• In some cases, the youth is actually the child of
  a parent in residential treatment. These must be
  handled separately.
• For this analysis, the data cases represent
  treatment episodes. A single individual may be
  represented multiple times.

7
Youth treatment episodes ranged from a low of
869 to a high of 1,198, an increase of 37.
Youth tx episode
Youth in tx with parent
8
Using the average length of stay, we can derive
a measure of overall system utilization.

• Average length of Stay (ALOS) varies
  significantly by treatment modality.
• We can derive a rough estimate of system
  Utilization by multiplying the number of
  treatment episodes by the average length of stay,
  by treatment modality and then summing over the
  totals.

9
With the exception of FY99-00, overall system
utilization generally increased and is now
(171,186) more than twice what it was in
FY1995-96 (81,500).
10

• What portion of the treatment resources are
  accounted for by referrals from the criminal
  justice system?

11
Sources of referrals to treatment have varied.
Referrals from mental health have declined, while
. self referrals and those from schools and
criminal justice appear to have increased.
12
As a proportion of all referrals, CJ referrals
have been relatively stable for three years at
about 40 of all youth AD tx episodes.
13
Utilization based on referrals from the CJ system
behaves somewhat like overall utilization.
14
CJ utilization as a proportion of total
utilization behaves much like the ratio for tx
episodes accounting for about 40 of tx
resources.
15

• How are treatment resources for
• CJ referrals distributed over gender,
  race/ethnicity, etc?

16
Youth referred from the CJ system were more
likely to be male, by a ratio of about 4 to 1,
which has remained relatively stable over time.
Female
Male
17
Referrals to treatment from the criminal justice
system by race/ethnicity.

• Among the CJ Referrals, African Americans and
  Native Americans are consistently
  over-represented in the youth treatment system.
• Since 1998, there has been a rapid increase in
  the number -- and percentage -- of episodes
  accounted for by African American youth.
• Asian American youth enrolled in publicly funded
  treatment only infrequently, and far below their
  proportion in either the general population
  (Census)or the juvenile justice (JJ) population.

18
The age distribution of referrals from the CJ
system has varied slightly over time. The overall
average age has changed by only about one-half
year 15.6 to 16.0.
19

• What were the substance abuse issues for clients
  referred from the CJ system at enrollment and how
  have the issues changed over time?

20
Use of marijuana among CJ referrals was more
widespread than was use of alcohol. Use of
alcohol appeared to be declining somewhat.
21
Use of heroin and cocaine, while relatively
limited, increased last year, and use of
amphetamines continues to affect approximately
30 of the CJ referrals.
22
Males referred from the CJ system were more apt
to report using marijuana, consistently over time.
23
Males referred from the CJ system were also more
apt to report using alcohol, again with
consistency, but with smaller difference.
24
But for amphetamines, females referred from the
CJ system were consistently more apt to report
use than were CJ-referred males.
25
Similarly for cocaine, female reported use
consistently exceeded male use among those
referred from the CJ system.
26
And similarly for opiates -- females referred
from the CJ system were more apt to report use
than were males.
27
Not surprisingly, (since the compose the single
largest racial/ethnic group), the substance-use
profile for Caucasian youth looks much like the
overall youth profile.
28
The profile for African-Americans looks quite
different -- evidencing little use of
amphetamines, cocaine or heroin, but nearly
universal use of marijuana.
29
The profile for Native Americans stands in
contrast to that for African Americans -- less
use of marijuana and alcohol, more use of
amphetamines, cocaine and heroin.
30
Hispanic youth appear to have shifted from
alcohol to marijuana. Sporadically, they report
use of the harder substances -- amphetamines,
cocaine and heroin.
31
Through the shifts in State policy, basic
referral patterns remained remarkably stable.
32
Of those who enrolled in treatment, what
percentage remained long enough to be engaged?
What percentage successfully completed the
treatment episode? On average,how long did a
successful client stay in treatment?

• Length of stay (LOS) varies by treatment
  modality. Thus, the length of time to
  engagement must also vary by modality
• AD Detoxification 5 days
• Residential care 30 days
• Outpatient care 90 days
• According to State OARs, successful completion
  involves completion of 2/3 of the treatment plan
  and remaining clean and sober for at least 30
  days.

33
The percentage of youth who engage in treatment
has varied over the years, without apparent
direction. The overall mean was about 30.
34
Engagement rates have varied substantially
depending on the treatment modality. Inpatient
engagement rates are generally higher than
outpatient rates, with the exception of DUII.
35
The distribution of engagement rates over gender,
race/ethnicity and age have varied over the years
without apparent consistency.

• In most cases, the differences were not
  statistically significant.
• In cases where the differences reached
  significance, the direction of the differences
  was not consistent from occasion to occasion.

36
Treatment completion rates vary substantially
over the treatment modalities. As in the adult
system, DUII completion rates are generally the
highest of all modalities. In-patient completion
rates generally exceed outpatient rates.
Beginning in FY 2000, residential treatment has
replaced CIRT.
37
The average length of stay (ALOS) for those who
completed treatment has varied substantially
over the years. In general, ALOS for outpatient
services exceeds that for inpatient care.
38
Reduction in use While abstinence is the
objective of treatment, it is important to
measure progress in the form of reductions in
substance use.

• CPMS measures reported usage of up to three
  substances at enrollment and at termination,
  using a seven point scale from 0 (none) to 6
  (More than 3 times per day).
• Reduction in use may be calculated by comparing
  use at enrollment with use at termination.

39
For the substance listed first, Those reporting
no use increased from 1 to 51. An additional
37 reported no change in their substance use
behaviors.
40
The pattern was similar for the second substance.
Those reporting abstinence grew from 5 to 58,
with 36 reporting no change in use behaviors.
41
The pattern was slightly different for the third
substance. More reported abstinence at enrollment
and nearly ¾ (73) reported abstinence at
termination. Less than ¼ (23) reported no change
in use behavior.
42
Treatment completion and use reduction are
strongly related. Both are related to length of
stay, but are statistically independent of client
background characteristics.

• Those who are judged to have successfully
  completed their treatment episode are more
  likely to have reduced their use and to have made
  larger reductions in use.
• Both treatment completion and use reduction are
  strongly related to the length of time clients
  remain in treatment. The longer the stay, the
  more likely to complete and to reduce use.
• Gender differences were inconsistent and not
  statistically significant.
• Ethnic/racial differences were also inconsistent,
  with some minority groups doing better and others
  doing worse compared with Caucasian youth.

43
Summary

• The number of youth AD tx episodes, including
  referrals from the CJ system, declined from
  1994-5 through 1997-8, but have increased again
  through last FY.
• CJ Referrals have consistently represented about
  40 of all episodes as well as overall treatment
  days (utilization).
• Males consistently outnumber females in youth AD
  tx by about 41.
• While males are more apt to report using alcohol
  and marijuana, females are more apt to report
  using the harder substances -- amphetamines,
  cocaine and heroin.

44
Summary (Cont.)

• Minority racial and ethnic groups were
  substantially over-represented in the publicly
  funded youth AD tx system -- with the exception
  of Asians.
• There has been a substantial increase in
  African-American youth in the AD tx system in
  the past three years.
• Despite their growth in the population, the
  number of Hispanic youth in the tx system on a
  referral from the CJ system has actually declined
  over the past three years.
• African-American youth used marijuana and, to a
  declining extent alcohol. However, few used
  amphetamines, cocaine or heroin.

45
Summary (Cont.)

• Treatment completion rates have varied over the
  years, but have generally been higher for
  inpatient episodes compared with outpatient
  episodes.
• Rates of reported abstinence increased, for the
  first two substances, from between 1 and 5 to
  over 50, and for the third substance from 26 to
  73.
• Of the factors measured, only the length of stay
  appears to effect treatment completion or use
  reduction.

46
Part 2Some insights based on the evaluation of
the Targeted Capacity Expansion grant (CSAT)

• Overview of the study
• 2. AD treatment population and the TCE client
  population
• 3. Characteristics of the TCE clients
• 4. Some lessons learned in the TCE project

47
Brief overview of the TCE project

• Beginning in August 1999, two dually qualified
  (AD and MH) clinicians began assessing DCJ youth
  for substance abuse problems.
• Youth were referred to the clinicians by Juvenile
  Court Counselors (JCC).
• Youth were referred for assessment based on JCCs
  perception of a substance abuse problem,
  particularly when the problem was associated with
  continued criminal behavior, and a history of
  failed attempts to reduce usage.
• Of particular interest were those youth perceived
  to have both substance abuse and mental health
  problems.

48
TCE Clients and the DCJ Treatment Population

• In the two Fiscal Years 1999-2000 and 2000-2001,
  there were 852 adolescent treatment episodes
  emanating from a criminal justice referral, of
  which 87 were outpatient episodes.
• During this same period, 503 youth were assessed
  by the TCE clinicians, resulting in 231 youth
  enrolling in treatment, some on multiple
  occasions.
• Thus, youth assessed under TCE represent
  approximately 30 (250/850) of all criminal
  justice enrollments in community treatment.
• Of the enrollments following the initial
  assessments, 90 were in outpatient services. At
  the final enrollment captured, the rate was
  approximately 87, mirroring the treatment
  population.

49
The proportion of females referred from the
criminal justice system has increased over the
past three years. As a whole,TCE appears to
accurately represent the gender distribution.
But, females appear to be under-represented among
those who enrolled in treatment.
50
The proportion of African-American youth referred
from the criminal justice system to treatment has
increased rapidly over the past three years. The
proportion in the TCE population was even higher
and the proportion who enrolled in treatment yet
again higher. Caucasians were consistently
under-represented.
51
The TCE project served youth over age 18 these
youth are generally excluded from the
adolescent treatment population. Youth aged 13
and under were under-represented in the TCE
population and among enrollees. Youth aged 16 and
17 were most apt to enroll in treatment.
52
The youth assessed under the TCE project and
referred to AD treatment were not a
representative sample of all DCJ referrals to
treatment.

• The TCE population includes more males and more
  non-Caucasian youth than does the general
  treatment population referred from DCJ.
• The TCE clients generally tend to be older than
  those referred into the adolescent treatment
  system, primarily because it includes youth 18
  years and older.
• Thus, the findings that follow may not be
  generalized to the entirety of DCJ referrals to
  AD treatment.

53
Characteristics of the TCE populationAlcohol and
drug diagnostics

• Over 85 of the DCJ youth assessed under TCE were
  assessed once the remaining 14 (n90) were
  assessed on multiple occasions.
• Of those who were assessed more than once, the
  diagnoses were consistent in only 39 of the
  cases. In 36, the last diagnosis was more severe
  than the initial diagnosis in 17 the last
  diagnosis was less severe.
• The average time between the initial and final
  assessments was approximately 7 months.
• In most cases, the several assessments were
  conducted by the same clinician.
• These data suggest that youth substance abusing
  behaviors are highly variable and diagnostics
  therefore unstable.

54
At the initial assessment, those who were
subsequently assessed again were less likely to
be diagnosed without a problem than were those
who were assessed only once. At the final
assessment, over 60 of those assessed on
multiple occasions were determined to be
dependent.
55
Using the final diagnosis, the differences
between males and females in their alcohol and
drug diagnoses were negligible.
56
The differences in diagnoses based on
race-ethnicity were also not large. However,
Caucasian youth were the most apt to be diagnosed
as dependent.
57
The mental health diagnoses for those assessed on
multiple occasions were similarly unstable.

• The mental health assessment was recorded as a
  simple yes/no dichotomy, with a third category
  for those for whom the diagnosis was not yet
  determined.
• Of the 90 youth assessed on multiple occasions,
  the mental health diagnoses were consistent
  between the first and last assessment for 46
  (51).
• For 38 youth (42), the diagnoses were
  inconsistent, equally divided between those who
  changed from Yes to No, and those who changed
  from No to Yes.
• As with the AD diagnoses, it appears that youth
  manifestation of mental health problems varies
  substantially over comparatively short periods of
  time.

58
Those who were assessed on multiple occasions
were slightly more likely to be diagnosed with a
mental health problem. Overall, more than
one-half of the DCJ clients assessed under the
TCE project were diagnosed with a mental health
problem.
59
The likelihood of a co-occurring mental health
problem increased with the severity of the
alcohol and drug diagnosis.
60
There was little difference in the distribution
of diagnoses based on gender.
61
There were slight differences in the distribution
of diagnoses based on race/ethnicity. While the
differences were not extreme, Hispanics were most
likely to be diagnosed with a substance abuse
problem only.
62
The percentage of clients diagnosed with both AD
and MH problems tended to decrease as client age
increased.
63
Client flow into and through treatment Only
about one-half the youth assessed were referred
to AD providers in the community. Over 40 were
referred elsewhere AITP, OYA, etc.
64
Females were slightly more likely to be referred
elsewhere, but the difference was not
statistically significant (p.11).
65
It appears that African-American youth were
disproportionately referred to treatment
providers other than AD providers located in the
community, compared with ALL other racial and
ethnic groups.
66
The likelihood of being referred to a community
treatment provider increased with client age. In
contrast, the likelihood of being referred
elsewhere decreased.
67
All of those not referred were diagnosed without
a problem. The likelihood of referral to
treatment other than a community provider
increased with the severity of the AD diagnosis.
68
The likelihood of referral increased with the
presence of mental health problems, particularly
the likelihood of referral to a provider other
than an AD community provider.
69
The nature of the referral varied depending on
the nature of the diagnosis. It also seems clear
that these diagnostics do not fully account for
the nature of the referral.
70
Enrollment rates were slightly higher among
males. (p 71
Rates of enrollment were slightly higher among
African-Americans and those of racial/ethnic
backgrounds other than three identified.
72
Rates of enrollment increased with age until age
16-17. They were actually lowest among the oldest
youth served under TCE.
73
Rates of enrollment in AD treatment increased
with increasing severity of the problem.
74
AD treatment enrollment rates were slightly
higher for those with mental health problems as
compared with those without.
75
Rates of enrollment in AD treatment were
highest, and nearly equivalent for those with a
diagnosed AD problem or with diagnosed
co-occurring disorders.
76
Female adolescents were slightly more likely to
engage in treatment than were their male
counterparts, but the difference was not
statistically significant.
77
Other demographic characteristics were, for the
most part, similarly unrelated to engagement
rates.

• The differences across racial/ethnic categories
  are small, except for those describing themselves
  as Hispanic/Latino.
• Engagement rates of Hispanics were at least 10
  lower than for the other racial and ethnic
  groups.
• Rates varied over the age groups, but without
  apparent direction.

78
The difference in engagement rates based on the
AD diagnosis are are greater than
79
are the comparable treatment engagement rates
based on the presence or absence of a mental
health problem.
80
Of the 179 final treatment episodes for which
termination data was available, 49 episodes (27)
were successfully completed.
81
Gender differences in the distribution of
termination status were not statistically
significant. However, no females failed to keep
their initial appointments and females were
slightly more likely to successfully complete
their final treatment episode.
82
The differences based on race/ethnicity were also
not statistically significant. However, Caucasian
youth were the least likely to successfully
complete their final treatment episode
African-American youth were most likely to have
terminated without prejudice.
83
The differences based on client age were also not
statistically significant and, for the most part,
without clear pattern. The likelihood of not
keeping the initial appointment did increase with
age.
84
The differences in likelihood of successfully
completing treatment across alcohol and drug
diagnostic categories do achieve statistical
significance (p 85
Youth with diagnosed mental health problems were
less likely to complete treatment than were those
without such problems.
86
Youth diagnosed with an alcohol and drug problem
only were more likely to complete treatment than
those with both AD and mental health problems.
87
Summary

• A substantial percentage of the youth referred to
  substance abuse treatment from the criminal
  justice system suffer from mental health as well
  as substance abuse problems.
• The likelihood of mental health problems
  increases with increasing severity of the alcohol
  and drug diagnosis.
• Substance abuse, but particularly mental health
  symptomology apparently varies over time with
  youth.
• The presence of co-occurring mental health
  problems did not appear to affect the likelihood
  of enrollment in substance abuse treatment
  provided by agencies located in the community.

88
Summary (Cont.)

• The presence of mental health problems did,
  however, appear to affect the likelihood of
  engaging in treatment and of successfully
  completing the treatment episode.
• While there were small differences in completion
  rates across gender, race/ethnicity and age, they
  were neither large nor entirely consistent.